I propose to enquire here into the way we endow our speech with meaning and into the way by which we make sense of speech that we hear spoken. I shall show that, notwithstanding their informal character, these acts possess a characteristic pattern, a pattern that I shall call the structure of tacit knowing; I shall show that to form such a structure is to create meaning. Both the way we endow our own utterances with meaning and our attribution of meaning to the utterances of others are acts of tacit knowing. They represent sense-giving and sense-reading with in the structure of tacit knowing. My enquiry shall outline the total structure of language, comprising both its formal patterns successfully established by modern linguistics and its informal semantic structure, studied so far mainly by philosophy.

It is frequently said that words like ‘now’, ‘then’, ‘ago’, ‘present’, ‘past’, ‘future’ and the various indications of tense, are ‘egocentric’ or ‘token-reflexive’ in character. I want to suggest, on the contrary, that the apparent egocentricity or token-reflexiveness of this class of expression is deceptive. It is perhaps not easy to see how on a point of this sort deception is possible, but a parallel case may make the position clearer.

Historically there have been two main freewill problems, the problem of freedom versus predestination, which is mainly theological, and the problem of freedom versus determinism, which has exercised the minds of many of the great modern philosophers. The latter problem is seldom stated in full detail, for its elements are taken as so obvious that they do not need to be stated. The problem is seen as an attempt to reconcile the belief in human freedom, which is essential if men are to be able to act morally, with determinism, the belief that every event is fully determined in all its details by the sum of its precedent causes. But even the meticulous Moore does not trouble to explore at length what is meant by determinism. He devotes one very short paragraph to the matter, and sums it up immediately afterwards as the view that ‘everything … has a cause’.

In his Philosophical Investigations, Wittgenstein introduces the notion of a ‘family resemblance’ to deal with certain problems. Talking of games and what they seem to have in common, he points out that there are no common features (or no common feature) in virtue of which we call all games ‘games’. Instead there are, he claims, many different similarities and relationships; he says ‘we see a complicated network of similarities overlapping and criss-crossing: sometimes overall similarities, sometimes similarities of detail’. (§ 66.) He then goes on to add: ‘I can think of no better expression to characterise these similarities than “family resemblances”; for the various resemblances between members of a family: build, features, colour of eyes, gait, temperament, etc. etc. overlap and criss-cross in the same way,—And I shall say: “games form a family”.’ (§ 67.) Wittgenstein also instances numbers as forming a ‘family’ in the same manner. This notion of a ‘family resemblance’ has come to be used by many philosophers to deal with a range of situations where there appears to be a difficulty in finding a single definite common property and yet there exists a desire to call some set of things by the same name. I myself have succumbed to this temptation. Perhaps the widest claim for the use of this device is that made by Mr Bambrough in an article entitled ‘Universals and Family Resemblances’ (Proceedings of the Aristotelian Society, 1960–61, pp. 207–222). This begins with the words ‘I believe that Wittgenstein solved what is known as “the problem of universals”’.